3DGS-CD: 3D Gaussian Splatting-Based Change Detection for Physical Object Rearrangement

作     者：Lu, Ziqi Ye, Jianbo Leonard, John 

作者机构：MIT Comp Sci & Artificial Intelligence Lab Cambridge MA 02139 USA 

出 版 物：《IEEE ROBOTICS AND AUTOMATION LETTERS》 (IEEE Robot. Autom.)

年 卷 期：2025年第10卷第3期

页      面：2662-2669页

核心收录：

学科分类：0808[工学-电气工程] 08[工学] 0811[工学-控制科学与工程] 

基　　金：ONR Neuro-Autonomy MURI [N00014-19-1-2571] ONR DURIP [N00014-23-12164] MIT Portugal Program 

主　　题：Three-dimensional displays Image segmentation Cameras Rendering (computer graphics) Training Object recognition Neural radiance field Solid modeling Visualization Point cloud compression Computer vision for automation mapping 

摘      要：We present 3DGS-CD, the first 3D Gaussian Splatting (3DGS)-based method for detecting physical object rearrangements in 3D scenes. Our approach estimates 3D object-level changes by comparing two sets of unaligned images taken at different times. Leveraging 3DGS s novel view rendering and EfficientSAM s zero-shot segmentation capabilities, we detect 2D object-level changes, which are then associated and fused across views to estimate 3D change masks and object transformations. Our method can accurately identify changes in cluttered environments using sparse (as few as one) post-change images within as little as 18 s. It does not rely on depth input, user instructions, pre-defined object classes, or object models - An object is recognized simply if it has been re-arranged. Our approach is evaluated on both public and self-collected real-world datasets, achieving up to 14% higher accuracy and three orders of magnitude faster performance compared to the state-of-the-art radiance-field-based change detection method. This significant performance boost enables a broad range of downstream applications, where we highlight three key use cases: object reconstruction, robot workspace reset, and 3DGS model update.